E-publish or perish

John Mcreery (jlm@TWICS.COM)
Tue, 14 Jun 1994 06:44:24 JST

"JOhn Mcreery suggests a technique for determining 'value' (my word, not
his) of publications by the "how-often-citen-criterion". The problem with
this criteria are many: The main Index (Social Sciences Citation Index)
only reveiws certain jounals/periodicals, not all; An author can cite
her/him self over and over again, thus increasing their own citations
frequencies; There is so much to read, that those with time limitations
tend to read in limited journal-circles, and thus cite from a limited
pool. I think a better measure (more subjective, less statistically
measurable, but hey, I'm a cultural anthropologist) is the quality and
relevance of the writing/argument/subject matter of the specific article.
In part this can be determined by reading the articles, and in part by
the kind of debate/response that the article generates (so partly
citation-frequencies, but also the quality/type of the responses).
Heather Y-L"

Don't want to be testy, but the one thing I expect our of a cultural
anthropologist is sensitivityto context. The notions I put on the table were
a response to Seeker1's call for some wayto bring greater prestige to
electronic publishing and the thought that in a market economy "quicker,
easier, more available," etc. usually mean "cheap." What people are griping
about is the value assigned to publication in "prestigeous" journals--
which, of course, is much like the value assigned to Rolex watches, BMWs,
Ivy-League educations and similar products that mayhave other intrinsic
merits but are clearly valued because they are rare. The (admittedly very
rough) suggestions I made were in the context of searching for a new
basis for "value" to add prestige to electronic publishing, a context in
which the computers managing the network could be programmed to keep
track of things like recommendations and cross-references automatically and
exclude self-references to help keep the game honest. With the kind of
graphics capabilities now becoming available, they could also be programmed
to image the connections at any level of detail desired to simplify
presentation of complex matters. "Hey, this is our candidate's activity
profile, see how it's grown in the last two years....[for example]."

Ideal? No. But wouldn't it be better to be able to say that x-thousand people
like what I wrote enough to recommend or cite it than be totally at the
mercy of the likes and dislikes of a few people (those nasty members of
the hiring/tenure committee) whose judgments of quality and relevance are
as idiosyncratic and, yes, politically motivated as any human being's are?

As things stand now, you can't get that kind ofinformation except bylaborious
methods that few people with other things to worryabout have time for.
We're talking the "Big Network," the "Cyberspace" or whatever else you
want to call it--I don't care. What matters is the possibilityof using the
the kind of data that might be
useful and make it widely and easily accessible.

Then comes the question, how to use it. That's a judgment about "value" and
that, of course, means politics or, let's be crude, marketing. What you
gonna offer the guys who're sitting pretty in the current system that will
make them interested in change? This scheme I'm suggesting looks like it
(1) would make their own jobs easier, (2) be fairer than the local politics
of small committees, (3) have the side-benefit of mapping intellectual
movements in a way that might be intrinsically interesting...All right it's
only a beginning. Over to you.

John McCreery (JLM@TWICS.COM)